WO2001049461A1

WO2001049461A1 - Edit device, edit method, and recorded medium

Info

Publication number: WO2001049461A1
Application number: PCT/JP2000/009417
Authority: WO
Inventors: Kyoko Furumura
Original assignee: Sony Corporation
Priority date: 1999-12-29
Filing date: 2000-12-28
Publication date: 2001-07-12
Also published as: US20020137425A1; US7063591B2; JP2001191281A; CN1124191C; KR20010102425A; CN1339997A; TW544614B

Abstract

When the user of a robot contacts with the robot in the same way, the behavior and growth of the robot are similar, causing the user to easily tire. The conventional problem is solved. An edit device and method of the invention is for displaying a predetermined behavior model of a robot according to which the robot behaves and for editing the displayed behavior model through a user's operation. Another edit device and method of the invention is for displaying a predetermined growth model of a robot according to which the robot grows and for editing the displayed growth model through a user's operation.

Description

Description Editing device, editing method and recording medium

The present invention relates to an editing device, an editing method, and a recording medium, and is suitably applied to, for example, an editing device, an editing method, and a recording medium for editing a growth model and / or a behavior model of a pet robot. Background art

In recent years, a four-legged walking robot has been proposed and developed by the applicant of the present invention, which performs actions in response to commands from the user and the surrounding environment. Such a pet robot has a shape similar to a dog or cat bred in a general household, and acts autonomously in response to a command from a user or the surrounding environment. In the following, a set of actions is defined and used as actions.

Also, this pet robot is equipped with a function to grow like a real dog or cat, giving the user a sense of closeness and satisfaction, and improving the amusement of the pet robot. Proposed by the applicant (Japanese Patent Application No. 11-1299276).

However, in the pet robot disclosed in Japanese Patent Application No. 11-1299276, the level of difficulty and complexity of the action and movement that appears (hereinafter referred to as the growth level) is set in stages. For example, if the user has completed “growing up” or the period until the next “growth” is long, the user may become tired of getting used to the behavior and movement of the pet robot. Was.

In addition, since such a pet robot determines its behavior and action by a predetermined program, for example, a command given by a user to a plurality of petrobots at the same growth stage and the surrounding environment are determined. If are the same, each such unit The behaviors and actions of the robot became the same, and it was difficult to bring up robots that perform user-specific actions and actions.

Therefore, in such a pet robot, if such a problem could be solved, it would be possible to prevent the user from getting tired and to further improve the amusement performance. Disclosure of the invention

The present invention has been made in view of the above points, and an object of the present invention is to propose an editing apparatus, an editing method, and a medium that can significantly improve amusement. In order to solve this problem, in the present invention, an editing device includes a display unit that visually displays the behavior model of a robot device that acts according to a predetermined behavior model, and edits the visually displayed behavior model in response to a user operation. An edit processing means for processing is provided.

As a result, in this editing device, the behavior model of the robot device is edited to be unique to the user, and the robot device is allowed to perform the behavior unique to the user, thereby providing the user with unique characteristics different from other robot devices. Therefore, it is possible to effectively prevent the robot device from getting tired of the action, and thus it is possible to realize an editing device capable of further improving the amusement property.

Further, in the present invention, in the editing apparatus, a display means for visually displaying the growth model of the robot apparatus which grows according to the predetermined growth model, and an editing processing for editing the visually displayed growth model in response to a user operation Means are provided.

As a result, in this editing device, the growth model of the robot device is edited to be unique to the user, and the robot device is allowed to grow unique to the user and to have a unique user characteristic different from other robot devices. Therefore, it is possible to effectively prevent the growth of the robot apparatus from getting tired, and thus to realize an editing apparatus that can further improve the amusement property. Further, in the present invention, in the editing method, a first step of visually displaying the behavior model of the robot device acting according to a predetermined behavior model, and editing the visually displayed behavior model according to a user operation And a second step to do this.

As a result, in this editing method, the behavior model of the robot device is edited to be unique to the user, and the robot device is caused to perform the behavior unique to the user, so that the robot device has unique features different from those of other robot devices. Therefore, it is possible to effectively prevent the behavior of the robot apparatus from getting tired, and thus to realize an editing method that can further improve the amusement property.

Further, in the present invention, in the editing method, a first step of visually displaying the growth model of the robot apparatus that grows according to the predetermined growth model, and editing the growth model that is visually displayed according to a user operation And a second step to do so.

As a result, in this editing method, the growth model of the robotic device is edited to be unique to the user, and the robotic device is made to grow unique to the user, and is provided with a unique user characteristic different from other robotic devices. Therefore, it is possible to effectively prevent the growth of the robot device from getting tired, and thus it is possible to realize an editing method capable of further improving the amusement property.

Further, in the present invention, an editing step includes a display step of visually displaying the behavior model of a robot apparatus that behaves according to a predetermined behavior model, and an editing step of editing the visually displayed behavior model in response to a user operation. The program for processing is stored in the recording medium.

As a result, in the program stored in the recording medium, the behavior model of the robot device is edited to be unique to the user, and the robot device performs the behavior unique to the user, and communicates with other robot devices. Can have unique features of different users, so that the behavior of the robot device can be effectively prevented from getting tired, and thus a recording medium that can further improve the amusement property. Can be realized. Further, in the present invention, an editing process including a display step of visually displaying the growth model of the robot device that grows according to a predetermined growth model, and an editing step of editing the growth model that is visually displayed according to a user operation. The program for performing is now stored in the recording medium.

As a result, the program stored in this recording medium edits the growth model of the robot device to be unique to the user, and allows the robot device to perform the growth unique to the user, and is different from other robot devices. Since it is possible to provide the features, it is possible to effectively prevent the growth of the robot apparatus from getting tired, and thus it is possible to realize a recording medium that can further improve the amusement property. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a perspective view showing an external configuration of a pet robot.

FIG. 2 is a block diagram showing a circuit configuration of the pet robot.

Figure 3 is a conceptual diagram showing a growth model.

FIG. 4 is a block diagram for explaining the processing of the controller.

FIG. 5 is a conceptual diagram for explaining data processing in the emotion / instinct model unit. FIG. 6 is a conceptual diagram showing a stochastic automaton.

FIG. 7 is a conceptual diagram showing a state transition table.

FIG. 8 is a conceptual diagram provided for explanation in a directed graph.

FIG. 9 is a conceptual diagram showing a first growth element list and a first growth element counter table.

FIG. 10 is a conceptual diagram showing a second growth element list and a second growth element counter table.

FIG. 11 is a block diagram showing the configuration of the editing device according to the present embodiment. FIG. 12 is a conceptual diagram showing a growth model editing screen.

FIG. 13 is a conceptual diagram showing an action model editing screen. FIG. 14 is a flowchart showing the editing processing procedure. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

(1) Pet robot configuration

(1-1) Schematic configuration of pet robot

In FIG. 1, reference numeral 1 denotes a pet robot as a whole, and leg units 3A to 3D are connected to the front, rear, left and right of the body unit 2, respectively, and the front and rear ends of the body unit 2 are respectively provided. The head unit 4 and the tail unit 5 are connected to each other.

In this case, as shown in FIG. 2, the body unit 2 includes a controller 10 for controlling the operation of the pet robot 1 as a whole, a battery 11 as a power source of the pet robot 1, and a battery sensor 1 2 And an internal sensor unit 15 including a heat sensor 13 and an acceleration sensor 14.

The head unit 4 also includes an external microphone comprising a microphone 16 corresponding to the “ears” of the pet robot 1, a CCD (Charge Coupled Dev ice) camera 17 corresponding to the “eyes”, and a touch sensor 18. A sensor section 19, a speaker 20 corresponding to a "mouth", and the like are arranged at predetermined positions.

In addition, the joints of the leg units 3A to 3D, the connecting portions of the leg units 3A to 3D and the body unit 2, the connecting portions of the head unit 4 and the body unit 2 and the like in the connecting portion of the tail part Yunitto 5 and the body portion Yunitto 2, their respective Akuchiyueta 2 1 i ~ 2 l _n are disposed.

The microphone 16 of the external sensor unit 19 collects command sounds such as “walk”, “down” or “follow the ball” given by the user via a sound commander (not shown) as a musical scale. The transmitted audio signal S 1 A is sent to the controller 10. Further, the CCD camera 17 captures an image of the surroundings, and sends the obtained image signal S 1 B to the controller 10. Further, as clearly shown in FIG. 1, the touch sensor 18 is provided on the upper part of the head unit 4 and detects the pressure received by a physical action such as “stroke” or “hit” from the user. The detection result is sent to the controller 10 as a pressure detection signal S 1 C.

Further, the battery sensor 12 of the internal sensor section 15 detects the remaining energy of the battery 11 and sends the detection result to the controller 10 as a remaining battery detection signal S2A. The temperature sensor 13 detects the temperature inside the pet robot 1 and sends the detection result to the controller 10 as a temperature detection signal S 2 B. Further, the acceleration sensor 14 detects acceleration in three axis directions (Z axis direction, Y axis direction and Z axis direction), and sends the detection result to the controller 10 as an acceleration detection signal S 2 C.

The controller 10 includes an audio signal S 1 A, an image signal S 1 B, a pressure detection signal S 1 C, and the like (hereinafter collectively referred to as an external information signal S 1) provided from the external sensor unit 19, Based on the remaining battery signal S 2 A, temperature detection signal S 2 B, acceleration detection signal S 2 C, etc. provided from the sensor unit 15 (hereinafter, these are collectively referred to as an internal information signal S 2), And the internal state, and the presence or absence of commands and actions from the user.

Then, the controller 1 0 includes a result of the determination, determines a subsequent action based on the control program stored in advance in the memory 1 OA, required Akuchiyue Ichita 21 on the basis of the determination result, to 2 l _n By driving the unit, the head unit 4 can be swung up, down, left and right, the tail unit 5A of the tail unit 5 can be moved, and each leg unit 3A to 3D can be driven to walk. Perform actions and actions.

Also, at this time, the controller 10 generates an audio signal S3 as necessary and gives it to the speaker 20, thereby outputting an audio based on the audio signal S3 to the outside, The LED (Light Emitting Diode) (not shown) placed at the position of the eye is turned off.

In this way, the pet robot 1 acts autonomously according to the external and internal conditions, the command from the user, the presence or absence of the action from the user, and the like. It has been made possible.

(1-2) Growth model of Pet Mouth Pot 1

Next, the growth function mounted on the pet robot 1 will be described.

In the case of this robot 1, the real animal grows in accordance with the history of operation input such as commands from the user and commands using the sound commander, as well as the history of his / her own actions and actions. They change their behavior and behavior as if they were.

That is, as shown in Fig. 3, the pet robot 1 is provided with four "growth stages" as growth processes: "childhood", "childhood", "adolescence", and "adult". I have. The memory 10A of the controller 10 stores four items of “walking state”, “motion”, “action” and “sound” for each of these “growth stages”. An action and action model including various control parameters, control programs, and the like that are the basis of the action and action are stored in advance.

Initially, the controller 10 follows the behavior and behavior model of “childhood”. For example, for “walking state”, the stride length is reduced and the “walking” is performed. Is a simple movement such as simply walking, standing, or sleeping, and a monotonous action such as repeating the same action. to, and as a like to "small short" cries reduce the amplification factor of the audio signal S 3 for "sound", controls each Akuchiyueta 2 1 i ~2 l _n and audio output.

At this time, the controller 10 includes a command input using a sound commander, a sensor input via the touch sensor 18 corresponding to “stroke” and “hitting”, and a determined number of successful actions and actions. Participates in reinforcement learning, sensor input via the touch sensor 18 that does not correspond to “stroke” or “hit”, and predetermined “growth” such as predetermined actions and actions such as “playing with a ball” For multiple elements (hereinafter referred to as growth elements), their occurrence is constantly monitored and counted. Then, the controller 10 determines each growth request based on the cumulative frequency of these growth factors. When the total value of the cumulative frequencies of the elements (hereinafter referred to as the total experience value of the growth factor) exceeds a preset threshold, the behavior and behavior model to be used is lower than that of the “childhood” behavior and behavior model. Change to “childhood” behavior and behavior model with high growth level. Then, the controller 10 follows the behavioral and behavioral model of the “childhood” and, for example, in the “walking state”, increases the rotation frequency of each actuator 21, to 21 „, etc. Refer to the previous action for "action" so that the movement becomes "slightly sophisticated and complicated" by increasing the number of movements, such as walking with "repression". In order to determine the next action, etc., the action has “a little purpose”, and for “sound”, extend the length of the audio signal and increase the amplification rate. as a longer large "bark, it controls the audio output from the Akuchiyueta 2 1 ~ 2 l _n and the speaker 2 0.

In the same manner, the controller 10 sets the behavior and behavior model each time the total experience value of the growth factor exceeds the threshold value set in advance corresponding to "adolescent" or "adult", respectively. sequentially changed more actions and behavior model of high growth level "adolescence" or "adulthood", the action and operation each according to the model Akuchi Yue Ichita 2 1, to 2 l _n rotational speed and the speaker 2 0 gradually raising the length and the amplification factor of the audio signal S 3 to give, each Akuchiyueta 2 1 in performing one operation, to vary the like rotational amount of to 2 l _n.

As a result, the pet robot 1 has an increased “growth stage” (that is, changes from “childhood” to “childhood”, “childhood” to “adolescence”, and “adolescence” to “adulthood”). According to the following, "walking state" power S "toddler" to "steady walking", "motion" from "simple" to "advanced / complex", and "action" from "monotone" to "action with purpose" And the "sound" changes gradually from "small and short" to "long and big".

In this way, in the robot 1, the input from the outside, “childhood”, “childhood”, “adolescence” and “adult It grows in four stages.

In this case, as apparent from FIG. 3, a plurality of behavioral and behavioral models are prepared for each of the “growth stages” of “childhood”, “adolescence” and “adulthood”.

In practice, for example, a behavior and behavior model of “childhood” is a behavior and behavior model (Ch ild 1) that performs a behavior and behavior of a rough and fast character, and the movement is smoother than this. A behavior and behavior model (Child 2) that performs behaviors and behaviors with a slow “unfriendly” character is provided.

In addition, as a behavior and behavior model of “adolescent”, a behavior and behavior model (Young 1) that performs behavior and behavior of an “irritated” personality that is rougher and faster than the “rough” personality of “boyhood”. ) And a behavior and behavior model (Young 2) that performs behaviors and movements with a “normal” character that is slower and smoother than this. A behavior and behavior model (Young 3) that performs behaviors and behaviors with the character “

In addition, the behavior and behavior model of "adult" is "aggressive", in which the movement is slower and faster than the "irritable" character of "adolescence", and it is difficult to perform the operation in response to commands from the user. Behavior and behavior model (Ad u1t1) that performs personal behavior and behavior, and “slightly violent” personal behavior that is smoother and slower than this and that is easy to perform in response to commands from the user And the behavior model (Adult 2), the behavior that is smoother and slower, the amount of action is smaller, and always performs the action in accordance with the command from the user. The behavior model (Adult 3), which is even slower, has a smaller amount of action, and always performs actions in response to commands from the user. And operation Model (Ad u 1 t 4).

Then, when raising the “growth stage”, the controller 1 ◦ selects one behavior and behavior model from among the behaviors and behavior models in the next “growth stage” based on the cumulative frequency of each growth factor. The behavior and behavior model to be used It has been changed to an operation model.

In this case, after the childhood, when moving to the next “growth stage”, the next “growth stage” action and behavior model that can transition from the current “growth stage” behavior and behavior model is determined. Only the transition between the behavior and the behavior model connected by the arrow in 3 can be performed. Therefore, for example, if the behavior and behavior model (C hild 1) that performs “violent” behavior and behavior in “childhood” is selected, “believable” behavior and It is not possible to make a transition to the action and action model (Young 3) that perform the action.

In this way, in this pet robot 1, as if a real animal forms its character according to the owner's breeding method, etc., the input from the user, the input history of the command, the own action and the operation history are obtained. Correspondingly, "personality" changes with "growth."

(2) Processing of controller 10

Next, a specific process of the controller 10 in the pet robot 1 will be described.

Functionally classifying the processing contents of the controller 10, as shown in FIG. 4, the state recognition mechanism unit 30 that recognizes the external and internal states, and the emotion based on the recognition result of the state recognition mechanism unit 30, And the emotions that determine the state of the instinct.The instinct model unit 31 and the recognition results and emotions of the state recognition mechanism unit 30.The action determination mechanism unit that determines the subsequent actions and actions based on the output of the instinct model unit 31. 3 2, an attitude transition mechanism 3 3 that makes a series of motion plans of the pet robot 1 for performing the action and action determined by the action determination mechanism 3 2, and an attitude transition mechanism 3 3 Actuator 2 1 based on the operation plan! 2 _n can be divided into a control mechanism section 34 for controlling the growth and a personality of the petro-pot 1.

Hereinafter, the state recognition mechanism section 30, the emotion / instinct model section 31, the action determination mechanism section 32, the posture transition mechanism section 33, the control mechanism section 34, and the growth control mechanism section 35 will be described in detail. . (2-1) Processing of state recognition mechanism section 30

The state recognition mechanism section 30 recognizes a specific state based on the external information signal S 1 provided from the external sensor section 19 (FIG. 2) and the internal information signal S 2 provided from the internal sensor section 15. Then, the recognition result is notified as state recognition information S10 to the emotion / instinct model unit 31 and the action determination mechanism unit 32.

In practice, the state recognition mechanism section 30 constantly monitors the audio signal S 1 A provided from the microphone 16 (FIG. 2) of the external sensor section 19 and as a spectrum of the audio signal S 1 A. When a spectrum with the same scale as the command sound output from the sound commander in response to a command such as "walk", "down" or "chase the ball" is detected, the command is recognized and given. The result is notified to the emotion and instinct model section 31 and the action decision mechanism section 32.

Further, the state recognition mechanism section 30 constantly monitors the image signal S 1 B provided from the CCD camera 17 (FIG. 2), and includes, for example, “a red circle” or “red circle” in an image based on the image signal S 1 B. When a plane that is perpendicular to the ground and higher than a predetermined height is detected, it recognizes that there is a ball and that there is a wall, and recognizes the recognition result as an emotional instinct model unit 31 and an action decision mechanism unit 3 Notify 2.

Further, the state recognition mechanism section 30 constantly monitors the pressure detection signal S 1 C provided from the touch sensor 18 (FIG. 2), and based on the pressure detection signal S 1 C, a signal having a predetermined threshold or more and a short time ( For example, when a pressure of less than 2 seconds is detected, it is recognized as “hit” (scored), and when a pressure of less than a predetermined threshold and for a long time (for example, 2 seconds or more) is detected, “stroke” (Praised) "and notifies the recognition result to the emotional instinct model unit 31 and the action decision mechanism unit 32.

On the other hand, the state recognition mechanism section 30 constantly monitors the acceleration detection signal S 2 C given from the acceleration sensor 14 (FIG. 2) of the internal sensor section 15, and for example, based on the acceleration signal S 2 C, When an acceleration equal to or higher than the set level is detected, it is recognized that "a large impact has been received". On the other hand, when an acceleration more than the gravitational acceleration is detected, "it has fallen (from a desk or the like)". Recognize and feel these recognition results Inform the instinct model 31 and the action decision mechanism 32.

Further, the state recognition mechanism section 30 constantly monitors the temperature detection signal S 2 B provided from the temperature sensor 13 (FIG. 2), and when a temperature equal to or higher than a predetermined value is detected based on the temperature detection signal S 2 B, The internal temperature has risen, "and the recognition result is notified to the emotion / instinct model unit 31 and the action determination mechanism unit 32.

(2-2) Emotions' Instinct Model Section 3 1 Processing

As shown in Fig. 5, emotions and instinct model section 31 are provided corresponding to the ^six emotions of "joy,""sadness,""surprise,""fear,""disgust," and "anger." Emotion unit 40 as a model of emotions, which corresponds to a basic emotion group 40 composed of 40A to 40F, and four desires of "appetite", "love", "search" and "exercise". The basic unit 4 1 consisting of the desire unit 4 1 A to 4 ID as the required unit model and the emotion unit 40 A to 40 F and each unit 4 1 A to 41 D And the intensity increase / decrease functions 42 A to 42 H provided.

Each emotion unit 40 A to 40 F expresses the degree of the corresponding emotion by, for example, the intensity up to the 0 to 100 level, and the intensity is obtained from the corresponding intensity increase / decrease function 42 A to 42 F. Based on the given intensity information S11A to S11F, the shape is changed every moment.

Each of the desire units 41 A to 41 D represents the degree of the corresponding desire by the intensity from the 0 to 100 level, similarly to the emotion unit 40 A to 40 F, and the intensity corresponds to the intensity. Intensity increasing / decreasing function The intensity increasing / decreasing function is changed every moment based on intensity information S12G to S12F given from 42G to 42K.

The emotion and instinct model 31 determines the state of emotion by combining the intensity of these emotion units 40A to 40F, and instinct by combining the intensity of these desire units 41A to 41D. Is determined, and the determined emotion and instinct state are output to the action determining mechanism 32 as emotion / instinct state information S12. The intensity increase / decrease functions 42A to 42G include the state recognition information S10 provided from the state recognition mechanism 30 and the current or past behavior of the pet robot 1 itself provided from the behavior determination mechanism 32 described later. Based on the action information S 13 to be represented, the intensity of each emotion unit 40 A to 40 F and each desire unit 41 A to 41 D is increased or decreased as described above in accordance with preset parameters. It is a function that generates and outputs intensity information S11A to S11G.

Thus, in the robot 1, the parameters of the intensity increase / decrease functions 42 A to 42 G are set to the respective behavior and behavior models (Baby 1, Child 1, Chi 1 d 2, Young 1 to Young 3, Ad By setting different values for each of the ult 1 to A du 1 t 4), the pet robot 1 can have a character such as “irritable” or “calm”.

(2-3) Processing of action decision mechanism 32

The action decision mechanism 3 2 corresponds to each action and action model (Baby 1, Chi i d 1, Chi i d 2, Young 1 to Young 3, Adult to Ad u 1 t 4), respectively. Thus, a plurality of behavior models are stored in the memory 1 OA.

The behavior determination mechanism 32 includes state recognition information 10 provided from the state recognition mechanism 30, the emotion units 40A to 40F and the desire units 41A to 4E of the emotion and instinct model unit 31. The next action or action is determined based on the 1D strength and the corresponding action model, and the determination result is output to the attitude transition mechanism section 33 and the growth control mechanism section 35 as action determination information S 14. .

In this case, the action determination mechanism part 3 2, as a method to determine the next action and motion, one node, as shown in FIG. 6 (state) ND _A. From connecting the same or any other node _ND _A0 ~ND Aπ each node or transition to ND _A0 to ND _An Aku AR _A. ~ Transition probabilities P set for AR _An respectively. Using an algorithm called probability Otomaton determining probabilistically based on to P _n.

More specifically, each node ND _A as an action model in memory 1 OA. ~ ND _An Each state transition table 50 as shown in FIG. 7 is stored, and the action determining mechanism 32 determines the next action or operation based on the state transition table 50.

Here in the state transition table 50, the node ND _A. In ND _An , input events (recognition results) as transition conditions are listed in order of priority in the “input event” line, and further conditions for that transition condition correspond to the “data name” and “data range” lines. Described in the column.

Therefore, in the node ND 100 defined in the state transition table 50 of FIG. 7, when a recognition result of “ball detected (BAL L)” is given, the “ball” of the ball given together with the recognition result is given. If the recognition result is given as “SIZE”, “0 to 1 000 range (0, 1 000)” or “OB STACLE”, the recognition result is given. The “distance (DI STANCE)” and “range of 0 to 100 (0, 100)” to the obstacle given along with is the condition for transition to another node.

Also this node ND _{1 ()} . Then, even when there is no input of the recognition result, the emotions that the action determination mechanism unit periodically refers to are the emotion units 40A to 40F of the instinct model unit 31 and the emotion units 40A to 40D of the desire units 41A to 41D. Among the strengths, the emotional unit 40 A to 40 F of “JOY”, “SUP RISE” or “SUDNE SS” has a strength of “100 to 100”. If it is in the range (50, 100) ", it can transition to another node.

Also in the state transition table 50, columns that node ND _A of "transition destination node" in the column of "probability of transition to another node". ~ ND _{An The} node names that can be transitioned from ND _An are listed, and all other conditions that can be transitioned when all the conditions described in each row of `` input event name '', `` data value '' and `` data range '' are met The transition probabilities of the nodes ND _A0 to ND _An are described in the row of “output action” in the column of “transition probabilities to other nodes”. Note that the sum of the transition probabilities of each row in the column of “transition probability to another node” is 100 [%]. Therefore, in the node NODE 100 in this example, for example, “ball is detected (BALL)”, and the “SIZE” force S of the ball is “range from 0 to 1 000 (0, 1 000)”. If the recognition result is given, it is possible to transit to “node NODE ₁₂ (node 1 20)” with a probability of “30 [%]”, and then the action or action of “ACT I ON 1” is output. Will be done.

And each behavior model, the node ND _A, which is described as such a state transition table 50, respectively. ~ ND _An is configured to connect several.

Thus, when the state recognition information S 10 is given from the state recognition mechanism 30 or when a certain period of time has passed since the last appearance of the action, the action determination mechanism section 32 stores the memory 10. The corresponding node ND _A of the corresponding behavior model stored in _A. Next, the next action or action (the action or action described in the row of “output action”) is stochastically determined using the state transition table 50 of ND _An , and the determination result is set as action command information S 14. and to output to the attitude transition mechanism part 3 3 and the growth control mechanism unit ^35. (2-4) Posture transition mechanism 3 3 Processing

When given the action determination information S14 from the action determination mechanism section 32, the posture transition mechanism section 33 performs a series of robot 1 operations for performing actions and actions based on the action determination information S14. And outputs the operation command information S15 based on the operation plan to the control mechanism unit.

In this case the posture transition mechanism unit 3 3, as a method to make a motion plan, each node ND _B by Unape Ttorobotto 1 can take attitude shown in FIG. To ND _B2 , and the transitionable node ND _B. Directed Aku AR _B representing the operation between ~ ND _B2. ~ One node ND _B tied with AR _B2 . Self-action operation to complete between ~ND _B2 arc AR _C. A directed graph expressed as ~ AR _C2 is used.

For this reason, the memory 10A stores in a database the starting point posture and the ending posture of all the motions that can be expressed by the pet port bot 1 that are the basis of such a directed graph (hereinafter referred to as network definition). The data is stored in the posture transition mechanism section 33 based on this network definition file. Generate directed graphs (not shown) for the body, head, legs, and tail.

And posture transition mechanism unit ₃ 3, "set up" from the action determining mechanism unit 3 2, "walk", "white your hands", "Yusure the head", the action command such as "moving the tail" is the action command Given as information S14, using the corresponding directed graph, a node associated with a specified posture or a specified operation from the current node while following the direction of a directed arc is associated. The control mechanism section searches for a path leading to a directed arc or a self-operating arc, and generates an operation command for sequentially performing an operation associated with each directed arc on the searched path as operation command information S15. Outputs one after the other to 3 and 4.

In addition, when an action command is given to the head, legs, or tail, the posture transition mechanism 33 changes the posture of the pet robot 1 based on the directed graph for the whole body according to the motion command. Return to one of the basic positions of “stand”, “sit”, “down”, and “station”, which is the position on the charging stand (not shown) for charging the battery 11 (FIG. 2). The operation command information S15 is output so as to change the posture of the head, the leg, or the tail using the directed graph of the corresponding head, leg, or tail.

(2-5) Processing of control mechanism section 3 4

The control mechanism section 34 generates a control signal S 16 based on the operation command information S 15 given from the attitude transition mechanism section 33, and based on the control signal S 16, each actuator 2 1 i by controlling the driving of the to 2 l _n, to perform the behavioral and operation specified in Bae Ttorobotto 1.

(2-6) Processing of growth control mechanism 35

Various states recognized based on the external information signal S1 and the internal information signal S2 from the state recognition mechanism section 30 are supplied to the growth control mechanism section 35 as state recognition information S20. The various states include, in addition to the specific states notified to the emotion / instinct model section 31 and the behavior determination mechanism section 32 as described above, for example, There is an input via the touch sensor 18 to an extent that does not correspond to “touch”.

In addition, the growth control mechanism section 35 includes the above-described growth element to be used as a reference element when raising the “growth stage” of various states based on the state recognition information S2 ° given from the state recognition mechanism section 30. A list such as that shown in Fig. 9 (A) described collectively (hereinafter referred to as the first growth factor list) 7 OA, and Fig. 9 (B) for counting the cumulative frequency of these growth factors, respectively. Such a counter table (hereinafter referred to as a first growth element counter table) 70 B is provided in the memory 10 A. Then, when given the state recognition information 20 from the state recognition mechanism section 30, the growth control mechanism section 35 determines whether or not the state obtained based on the state recognition information S20 is a growth element. Judgment is made based on the growth element list 7 OA of the above, and if the state is a growth element, the corresponding count value (experience value) in the first growth element counter table 70 B is increased by one.

In addition, the growth control mechanism section 35 should be a reference element when raising the “growth stage” among the actions obtained based on the action command information S 14 given from the action determination mechanism section 32 as described above. A list as shown in Fig. 10 (A), which summarizes the above-mentioned growth factors (hereinafter referred to as a second growth factor list) 7 1 A, and counts the cumulative frequency of these growth factors A counter table (hereinafter, referred to as a second growth element counter table) 71 B as shown in FIG. 10B is provided in the memory 10A.

Then, when given the action command information S14 from the action determination mechanism section 32, the growth control mechanism section 35 determines whether the action or action obtained based on the action command information S14 is a growth element. Judgment is made based on the second growth factor list 71A, and if the action is a growth factor, one corresponding count value (experience value) in the second growth factor counter table 71B is determined. increase.

Furthermore, as described above, when the count value in the first or second growth element counter table 70B or 71B is increased, the first and second growth element counters are increased. Increase the "growth stage" prepared separately from tables 70B and 7IB The count value of the counter for judging whether or not to grow or not (hereinafter referred to as the total growth experience value counter) is increased by one. It is determined whether or not a count value set in advance as the end condition of the “step” has been reached.

Then, when the count value of the total growth experience value counter reaches the count value set in advance as the end condition of the current “growth stage”, the growth control mechanism unit 35 changes the behavior and the operation model to the next “ Which action and behavior model in the `` growth stage '' should be changed is determined based on each count value in the first and second growth factor counter tables 70 B and 71 B, and the determination result is changed command information S As 22, it notifies the emotion and instinct model section 31, the action determination mechanism section 32 and the posture transition mechanism section 33.

As a result, based on the change command information S22, the emotion and instinct model unit 31 specifies the parameters of the intensity increase / decrease functions 42A to 42G described above with reference to FIG. Change to model value. The action determining mechanism 32 changes the action model to be used to the specified action and action model based on the change command information S22. Further, based on the change command information S22, the posture transition mechanism section 33 then selects one of the directed arcs and self-operation arcs from the directed arcs and self-operation arcs corresponding to a plurality of behavior and operation models. When the arc must be selected, change the setting so that the directed arc of the specified action and behavior model / self-action arc is selected.

As can be seen from this, the behavioral and behavioral models are the emotions corresponding to the "character" in the "growth stage" .The parameters of the intensity increase / decrease functions in the instinct model part 31 1 4 2 A to 4 2 G It consists of a value, an action model in the action determination mechanism section 32, a directed arc and a self-action arc in the attitude transition mechanism section 33, and the like.

In this way, the growth control mechanism section 35 controls “growth” according to the history of operation inputs such as commands from the user and commands using the sound commander, and the behavior and operation history of the user. (3) Editing device according to the present embodiment

(3-1) Configuration of editing device

In FIG. 11, reference numeral 100 denotes an editing device according to the present embodiment, which includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (R and om Accessory). (Memory) 103, display processing circuit 104, media unit 105, interface circuit 106 and SCS I (Small Computer)

An interface unit 107 is configured by connecting the interface circuit 107 via the CPU bus 108, and an operation unit 1 including a keyboard mouse and the like via the interface circuit 107. 09, and to the hard disk drive 110 via the SCSI interface circuit 107.

Further, the display processing circuit 104 is connected to the monitor 111, and the media unit 105 is inserted into a semiconductor memory, for example, a memory stick 112, so that the media unit 111 is inserted. Information is written to and read from the inserted memory stick 1 12 by a driver (not shown) in 05.

In this case, the CPU 101 reads out the application program stored in the ROM 102 to the RAM 103, reads out the image data in the ROM 102 based on the application program, and displays this in the display processing circuit 1. By giving the video signal S100 to the monitor 111 via the 04, an initial screen (not shown) based on the image data is displayed on a display unit (not shown) of the monitor 111.

In response to the initial screen displayed on the display unit of the monitor 111, the CPU 101 operates the media unit 1 corresponding to the memory stick 112 by the user.

When 05 is entered, for example, various information D100 of the robot 1 set in advance is read from the memory stick 111, and corresponding image data is read from the ROM 102 based on the information. The following 後述 based on the image data A GUI (Graphical User Interface) screen for editing the program of the growth and / or behavior model of the robot 1 is displayed on the display unit of the monitor 111.

Further, in the state where the GUI screen is displayed on the monitor 111 as described above, the CPU 101 responds to the command S 101 supplied from the operation unit 109 via the interface circuit 106 by the user. The screen operated based on this, and the numbers and characters input from the operation unit 109 via the interface circuit 106 are displayed on the display unit of the monitor 111 via the display processing circuit 104. Is superimposed on the GUI screen displayed.

Then, the CPU 101 sequentially edits the growth model and the behavior model of the pet robot 1 according to the operation of the user, and then obtains the edited result as edited data D 101, and the SCS I interface. A backup is made to the hard disk device 110 via the circuit 107, and the edited data D101 is stored in the inserted memory stick 112 via the driver of the media unit 105. . In this way, the editing device 100 can execute the program of the growth and / or behavior model of the robot 1 preset on the memory stick 112 inserted in the media unit 105 according to the user's preference. So that they can be edited.

(3-2) Configuration of GUI screen displayed on monitor 1 1 1

Actually, among the GUI screens displayed on the display unit of the monitor 1 1 1, the growth model edit screen GU 1, which is a graph as shown in Fig. 12, is a set stored in the memory stick 1 1 2 Robot 1's “childhood” behavior model (behavior model A), “childhood” behavior model (behavior model Bl, B2), and “adolescent” behavior model (behavior model Cl, C2 , C 3), etc., a growth model display area GU 1 A for displaying a growth model, and a plurality of types of editing items PA 1, PA 2, and so on for editing the growth model of the growth model display area GU 1 A. It consists of an item display area GU1B consisting of PA3A, PA4 and PA5. In addition, various editing items PA 1 to PA 5 in the item display area GU 1 B include commands given by the user to the pet robot 1 (FIG. 1) and icons I corresponding to the behavior of the pet robot 1 itself. 1! A ~ I 1 _5, icon I 2 is made to correspond to the behavior and operation pet robot 1 performs contrast, ~ I 2 _s and icons I ~ I 3 made to correspond to the part for forming the growth model and a _5, consisting of pet icons to correspond to the emotion of the robot 1 and instinct I 4 i ~ I 4 ₆ and I 5 ₁ ~ I 5 ₄ like.

By operating the growth model of the growth model display area GU 1 A on the growth model editing screen GU 1 using a desired icon of the item display area GU 1 B selected by the user, for example, “Infant” When the emotion of “joy” of the pet robot 1 reaches 100% in the behavior model A of the “period”, the “growth stage” of the pet robot 1 is changed to the behavior model B 1 of the “childhood”. ,

When the emotion of “anger” reaches 100 [%], the “growth stage” of the pet robot 1 transitions to the behavior model B 2 of “childhood”, and the instinct of “appetite” becomes 100 [%]. ), The “growth stage” of the pet robot 1 is viewed through the relay node (mouth) and the “favorite color” is viewed 50 times, and a transition is made to the behavior model B 2 of “childhood” You can edit it to

When the pet robot 1 performs “walking” 50 times in the “boyhood” behavior model B 1, the “growth stage” of the pet robot 1 transitions to the “adolescent” behavior model C 1, and “ By performing the action of “play with ball” 50 times, the “growth stage” of the pet robot 1 can be edited so as to transition to the “adolescent” behavior model C 2, and the “boyhood” behavior model B 2 When the pet robot 1 performs the action of “playing with po / le” 50 times, the “growth stage” of the pet robot 1 is shifted to the behavior model C 2 of “adolescent” and the “loud sound” is When heard 0 times, the pet robot 1 can be edited so that the "growth stage" of the pet robot 1 is shifted to the behavior model C3 of "adolescence".

In this way, in the editing device 100, the growth model editing screen GU 1 displayed on the monitor 111 is operated by the user, so that the pet robot 1 The growth model program can be edited according to the user's preference. The behavior model editing screen GU2, which is a graph displayed on the display unit of the monitor 111, has a memory, as shown in FIG. A part of the behavior model A selected by the user from the behavior model A, Bl, B2, and C1 to C3 of the robot 1 stored in the stick 1 1 2 is displayed. Multiple editing items PA1, PA2, PA3B for editing a part of the behavior model of the editorial area GU2A, which is almost the same as the item display area GU1B described above. , 4 and? It consists of an item display area GU 2 B consisting of A 5.

In addition, among the various editing items PA1 to PA5 of the item display area GU2B, the editing item PA3B is an icon I3B corresponding to a part for forming the behavior model A selected by the user. , ~ I 3 B ₅ etc. Then, the behavior model of the editing area GU 2 A on the behavior model editing screen GU 2 is operated by using a desired icon of the item display area GU 2 B selected by the user. When the pet robot 1 hears a “loud sound” in the node ND _C1 , the node moves forward in the direction indicated by the directed arc AR _C1A1 with a transition probability of 50%, and the relay node PAR, to move, after this "give voice", the flow proceeds in the direction indicated by the directed arc AR _C1A2 taking action, via a relay node PAR ₂ caused a "stand" action, to proceed in the direction shown by the directed arc AR _C1A3 _Thereby , the behavior of the self- _moving arc AR _C1A indicated by a dashed line consisting of these directed arcs AR _{C1A1 to} AR C1A3 can be edited.

Also, if a “ball” exists within 50 cm from the pet robot 1 at the node ND _C1 of the behavioral model A, the _{robot moves} in the direction indicated by the directed arc AR _C1B1 to the relay node PAR ₃ and then “ walk "taking action proceed in the direction indicated by a directed arc AR _C1B2, via a relay node PAR ₄ to take action to" kick the pole ", by proceeding in the direction indicated by the Yes Koaichiku AR _C1B3, these The behavior of the self- _moving arc AR _C1B shown by the dashed line consisting of the directed arcs AR _{C1B1 to} AR _C1B3 can be edited. You.

Via the relay node PAR ₅ proceeds in the direction indicated by the directed arc AR _C21 transition probability Similarly hear behavior model A node ND _C1 Nioitepe' preparative robot 1 force S "loud sound" of the 50 [%] `` Sit down '' and transition to node ND _C2 , and when the instinct of “Appetite” of pet robot 1 reaches 50 [%] in node ND _C1 of behavior model A, transition to node ND _C3 You can edit it to

Incidentally, at this time, these icons are selectively moved from the item display area GU2B to a desired part in the behavior model of the edit area GU2A by a so-called drag-and-drop operation by the user. Has been made. At this time, each icon can be selected any number of times.

In this way, in the editing apparatus 100, by operating the action model editing screen GU 2 displayed on the monitor 111 by the user, the program of the action model of the pet robot 1 can be changed according to the user's preference. Can be edited

(3-3) Editing procedure of CPU 101

Here, in practice, when the power is turned on by the user, the CPU 101 of the editing apparatus 100 grows and grows the above-described robot 1 according to the editing processing procedure RT1 shown in FIG. Alternatively, the program of the behavior model is edited.

That is, when the power is turned on by the user, the CPU 101 starts the editing processing procedure RT1 in step SP0, and in the following step SP1, inserts a “Memory Stick” into the display section of the monitor 111. Please display a message saying "Please" as an initial screen (not shown).

Thereafter, the CPU 101 proceeds to the next step SP2, and when the user inserts the memory stick 112 into the media unit 105 of the editing device 100, the CPU 101 pre-starts from the memory stick 112. The various types of information D100 of the set robot 1 are read, and the corresponding image data is read from the ROM 102 based on the information, and the growth and behavior model of the robot 1 based on the image data is read. Display the growth model edit screens GU 1 and GU 2 (Figures 1 and 2) for editing Dell on the monitor 1 1 1 display unit.

Then, the CPU 101 proceeds to the following step SP 3 and proceeds to the next step SP 4 when the user selects to edit the growth model on the growth model editing screen GU 1 displayed on the monitor 111. When the growth model is edited and the user selects a desired behavior model from the growth model editing screen GU1 in step SP3, the screen displayed on the monitor 1 1 1 is changed to the selected behavior model. The screen is switched to the corresponding behavior model editing screen GU2, and the process proceeds to the next step SP4, where the selected behavior model is edited.

In step SP4, the CPU 101 responds to the user operating the growth model editing screen GU1 or GU2 displayed on the monitor 111 via the operation unit 109. Edit the growth or behavior model of the pet robot 1 corresponding to the growth model edit screen GU1 or GU2.

Then, the CPU 101 proceeds to the subsequent step SP5, determines whether or not to end the editing of the growth or behavior model of the pet robot 1 corresponding to the growth model editing screen GU1 or GU2, and If an instruction to edit another growth or behavior model of the pet robot 1 is given as a command S101, a negative result is obtained, the process returns to step SP3, and the growth model editing screen GU is displayed again on the monitor 111. Display 1 to let the user select the desired growth or behavior model.

Thereafter, the CPU 101 executes the following steps SP 4 and SP 5 in the same manner as described above, and loops the steps SP 3—SP 4—SP 5—SP 3 until a positive result is obtained in step SP 5. Repeatedly, in step SP5, when the user gives an instruction to complete the growth of the pet robot 1 and edit the behavior model as the command S101, a positive result is obtained, and the process proceeds to the next step SP6. In step SP6, the CPU 101 backs up the edited result as edited data D101 to the hard disk device 110 via the SCS I interface circuit 107. And proceed to step SP 7 The edited data D101 is stored in the inserted Memory Stick 112 via the driver of the media unit 105.

Then, when the CPU 101 finishes storing the edited data D101 on the memory stick 112 in step SP7, the CPU 101 proceeds to step SP8, and says, "Remove the memory stick," on the monitor 111. Message (not shown) is displayed, and when the user removes the memory stick 112 from the media unit 105 by the user, the process proceeds to the next step SP 9 to end the editing process procedure RT 1 I do.

In this way, in the editing processing procedure RT1, the program for the growth and / or behavior model of the robot 1 preset on the memory stick 112 inserted into the media unit 105 is It can be edited according to the user's preference.

(4) Operation and effect of this embodiment

In the above configuration, in the editing device 100, the user performs actions such as “hit” and “stroke” performed by the user on the pet robot 1, a command given by the user using the sound commander, and the pet robot 1. editing according to the preference of the user graded the pair Ttorobo' sheet 1 of growth and or behavioral model program in accordance with actions you play with first force ^s "ball".

Therefore, the editing device 100 edits the personality of the pet robot 1 to be unique to the user, and causes the pet robot 1 to grow and behave in a manner unique to the user, so that the user is different from other pet robots. It can be provided with unique features, thus making it harder for the user to get tired and giving the user a greater sense of intimacy and satisfaction.

According to the above configuration, by editing the program of the growth and / or behavior model of the robot 1 according to the user's preference, the robot 1 can be provided with a unique feature of the user. Can make the user hard to get tired of and give the user a more intimate and satisfying feeling. An editing device 100 that can significantly improve the performance can be realized.

(5) Other embodiments

Note that, in the above-described embodiment, a case has been described in which the present invention is applied to the editing device 100 that edits the behavior and behavior model of the pet robot 1. However, the present invention is not limited to this. For example, the present invention is applied to a normal personal computer using a recording medium on which an application program for editing the behavior and behavior model of the robot 1 in such an editing device 100 is recorded. The behavior and behavior model of the pet robot 1 and the like may be edited, and can be widely applied to various other editing devices. In the above-described embodiment, a case has been described in which the object of editing of the present invention is applied to the growth and behavior model of a quadruped walking type robot 1 configured as shown in FIG. However, the present invention is not limited to this, and can be widely applied to robot devices having various other configurations. It can also be applied to characters moving on a monitor screen in computer graphics.

In this case, in the embodiment described above, the actions and or motion generating means for generating a behavior and motion model behavior and operation based on the controller 1 0, § Kuchiyueta 2 1, to 2 l _n, the speaker 2 Although the description has been given of the case where the light emitting device is configured by the LED at the position of 0 and the position of the “eye”, the present invention is not limited to this, and various other structures depending on the form of the robot mounting to which the present invention is applied. Furthermore, in the above-described embodiment, the case has been described where the personality and the growth level are changed based on both the input history from the outside and the own action and operation history. However, the present invention is not limited to this. Based on only one of the external input history and its own behavior and operation history, or based on the external input history and its own behavior and operation history, By taking into account the original, it is also possible to change the character and growth level Bae Ttorobo' door 1 at the timing other than "growth". Furthermore, personality and growth level are changed based on either one's own action history and action history. Is also good.

Further, in the above-described embodiment, the case where the pet robot 1 is made to “grow” stepwise has been described. However, the present invention is not limited to this, and it is possible to detect the state of the growth element, The value of the control parameter may be changed successively each time the action or action is performed, so that the "growth" is performed steplessly.

Further, in the above-described embodiment, a case has been described where the pet robot 1 is allowed to “grow” in four stages of “childhood”, “childhood”, “adolescence”, and “adult”. However, the present invention is not limited to this, and the number of “growth stages” may be set to a number other than four.

Further, in the above-described embodiment, a history such as a contact input via the touch sensor 18, an image captured by the CCD camera 17, and a command sound input using a sound commander is applied as an external input history. However, the present invention is not limited to this, and an input history from the outside in addition to or in addition to these may be applied.

Furthermore, in the above-described embodiment, a case has been described in which a plurality of behavior and behavior models are prepared for each of the “growth stages” after the “childhood”. However, the present invention is not limited to this. It is good to prepare only one behavior and behavior model for each “growth stage”.

Further, in the above-described embodiment, a case has been described in which four items of “walking state”, “motion”, “action”, and “sound” are changed according to “growth”. However, the present invention is not limited to this, and other items may be changed with “growth”. Industrial applicability

INDUSTRIAL APPLICABILITY The present invention can be applied to, for example, an editing device for editing a growth model or an action model of a pet robot.

Claims

The scope of the claims

1. Display means for visually displaying the behavior model of a robot device acting according to a predetermined behavior model,

Editing processing means for editing the action model visually displayed in response to a user operation;

An editing device comprising:

2. The above display means

Visually display the behavior model of the robot device as a graph

3. The editing device according to claim 1, wherein

3. display means for visually displaying the growth model of the robot device growing according to a predetermined growth model;

Editing processing means for editing the growth model visually displayed in response to a user operation;

An editing device comprising:

4. The above display means

Visually display the growth model of the robot device as a graph

4. The editing device according to claim 3, wherein:

5. a first step of visually displaying the behavior model of the robot device acting in accordance with the predetermined behavior model;

A second step for editing the above-mentioned behavior model displayed in response to a user operation;

Editing method characterized by comprising.

6. In the first step above,

Visually display the behavior model of the robot device as a graph

6. The editing method according to claim 5, wherein:

7. a first step of visually displaying the growth model of the robotic device growing according to the predetermined growth model;

A second step of editing and processing the visible growth model in response to a user operation.

Editing method characterized by comprising.

8. In the first step above,

Visually display the growth model of the robot device as a graph

8. The editing method according to claim 7, wherein:

9. A display step of visually displaying the behavior model of the robot device acting according to the predetermined behavior model;

A program for performing an editing process including an editing step of performing an editing process on the behavior model visually displayed in accordance with a user operation is stored.

A recording medium characterized by the above-mentioned.

1 0. In the above display step,

Visually display the behavior model of the robot device as a graph

10. The recording medium according to claim 9, wherein:

1 1. a display step of visually displaying the growth model of the robot apparatus growing according to a predetermined growth model;

An editing step for editing the visible growth model in response to a user operation. A recording medium storing a program for performing an editing process having a step.

1 2. In the above display step,

12. The recording medium according to claim 11, wherein the growth model of the robot device is visually displayed as a graph.